Crucial BX100 1TB SSD Review

[Guest]

The MX200 will eventually replace its MX100, while the BX100 series is a new low-cost drive designed to replace your hard drive.

Crucial BX100 1TB SSD Review : Read more

 

[salgado18]

My only issue with the drive is endurance. Because a secondary drive for storing data is meant to be the safe repository, wouldn't a hard drive be a better match, even with the lower performance?

 

[logainofhades]

You can currently get an 850 evo for less than a BX 100. http://pcpartpicker.com/parts/compare/crucial-internal-hard-drive-ct1000bx100ssd1%2Csamsung-internal-hard-drive-mz75e1t0bam/

 

[MrMusAddict]

"As a subsidiary of Lexar, parent company of Micron, Crucial has access to the highest-quality flash before competing drive vendors can buy it up."

You have it backwards there. It should read:

"As a subsidiary of Micron, parent company of Lexar..."

 

[Chris Droste]

i Love my MX100 512GB but i dunno if I'd go anything cheaper than that unless it's older/proven like a Samsung 830 or a Crucial m.4. i mean, $210 for 512GB is really damn sweet for a +90k IOps drive

 

[JPNpower]

Silicon motion vs Marvell. Interesting..... Would it dethrone the OCZ Arc100?

 

[dachiesa]

I grabbed a $85 sale for the 256GB one and upgraded my 120GB Kingston SSDNOW to this and I am very happy to have that extra space (the 120GB Kingston was 60 when I bought it last summer)

 

[soldier44]

Time to move to a 1Tb SSD from a 256gb one I have now, this may be the one get.

 

[mczak1]

The last page is saying "The cheaper controller is missing advanced features like DevSlp support and hardware-based encryption" - that is however untrue as far as the encryption is concerned (look it up on SiliconMotions site). I guess though Crucial decided to not make it available so there's some feature difference to the MX series. There could of course be some problems with it as well, but other SSDs based on the same controller offer hardware based encryption.

 

[mapesdhs]

Re the 128KB Sequential Write Performance graph - please stop using graphs
that do not have a zero origin. They are thoroughly visually misleading. What's
the point of using a graph at all if the bars' relative sizes are so different because
of a non-zero origin? The line graph makes it obvious the BX100 is slower but
not that much slower than the upper end models, whereas the bar graph at
first looks far worse. Also, the other graphs have not been done in the same way.

Best not to use non-zero-origin graphs at all.

Ian.

 

[SinxarKnights]

Time to move to a 1Tb SSD from a 256gb one I have now, this may be the one get.

I got the MX100 256GB and simply love it. I would like to get this and just use it for storage. I mean it would last pretty much forever since it isn't mechanical and the data isn't constantly being written to it.

 

[damric]

Chris,

What happens when you hard reset this SSD, say...50 times. Does it brick like the MX100s do?

Just wondering, since now we are seeing OCZ SSDs with a capacitor built-in for loss of power.

 

[JPNpower]

Chris,

What happens when you hard reset this SSD, say...50 times. Does it brick like the MX100s do?

Just wondering, since now we are seeing OCZ SSDs with a capacitor built-in for loss of power.

Yeah Chris, I thought we established that this should be a crucial (lol pun) part of all SSD tests.

 

[laststop311]

Since the 1TB 850 evo with TLC V-Nand is the same price and a better quality drive in every way, faster sequential, faster random, better consistency, higher endurance. the BX100 drive would have to be priced more towards 300 dollars to make it a better buy over the 850 evo.

 

[laststop311]

Time to move to a 1Tb SSD from a 256gb one I have now, this may be the one get.

It's not the 1 to get the samsung 850 evo 1TB size is almost identical in price and has top tier performance compared to this. Or for the very same price you can get a 400GB intel 750 pci-e 3.0 x4 nvme drive that is 4x faster than even the very best sata III based ssd. For many people 400GB of 5-6x faster speed than budget 1TB SATA III drives is more important than the capacity for the same price. The intel 750 finally shifts the bottleneck back to the cpu.

 

[KaiserPhantasma]

put a 1TB SSD at a $100 price range please

 

[riderrigo]

I'm not paying $400 for a hard drive where I won't access many of the files for a long time. I'm still going to run the $100 SSD with various regular HDDs setup regardless.

 

[JPNpower]

Since the 1TB 850 evo with TLC V-Nand is the same price and a better quality drive in every way, faster sequential, faster random, better consistency, higher endurance. the BX100 drive would have to be priced more towards 300 dollars to make it a better buy over the 850 evo.

...... You do know that TLC is a bad thing (comparatively) right?

 

[unityole]

Since the 1TB 850 evo with TLC V-Nand is the same price and a better quality drive in every way, faster sequential, faster random, better consistency, higher endurance. the BX100 drive would have to be priced more towards 300 dollars to make it a better buy over the 850 evo.

those numbers are masked and shows higher due to overclocked controller and slc cache. whjich is why 840 sucks and i aint buying a 850 until months later confirmed there are no issue, but then at that time new ssds are gonna come out so prob just skip 850 all together.

 

[nitrium]

"We want to focus mostly on the QD4 to QD8 range for typical consumer workloads."

Typical consumer workloads are actually QDs 1 to 4.

 

[boletus]

Re the 128KB Sequential Write Performance graph - please stop using graphs that do not have a zero origin. They are thoroughly visually misleading. What's the point of using a graph at all if the bars' relative sizes are so different because of a non-zero origin? The line graph makes it obvious the BX100 is slower but not that much slower than the upper end models, whereas the bar graph at
first looks far worse. Also, the other graphs have not been done in the same way.

Best not to use non-zero-origin graphs at all. Ian.

Enabled scripts and logged in just to upvote that comment.

 

[CRamseyer]

The chart issue is my fault. I didn't catch it when I built the charts. I'll have it fixed for the next one.

When looking at mixed workloads you want to look a little higher up the scale. There isn't a QD1 in mixed workloads because you have to start with 1 read and 1 write, so QD2. It really depends on your workload but I tends to look in the 4 to 8 range rather than just 2 and 4.

 

[nitrium]

When looking at mixed workloads you want to look a little higher up the scale. There isn't a QD1 in mixed workloads because you have to start with 1 read and 1 write, so QD2. It really depends on your workload but I tends to look in the 4 to 8 range rather than just 2 and 4.

From this very website for booting of Win 7:
"The boot-up processes that we monitored typically consist of approximately 88% random read access and 12% sequential read access, with an average queue depth of 1.4."
http://www.tomshardware.com/reviews/momentus-xt-750gb-review,3223-5.html
For loading CoD: MW3:
"We found that the workload consists of approximately 18% random access reads and 82% sequential access reads, which, by coincidence, is almost the inverse of what we saw during the boot-up process. The average block transfer size increases to 79 KiB and the average queue depth is one."
http://www.tomshardware.com/reviews/momentus-xt-750gb-review,3223-6.html

Actually, an interesting SSD article might be to do further analysis of this with many more applications (e.g. Office, Photoshop, various games, web browsing etc). I suspect you'll find QDs of 1 to 2 are by far and away a "typical consumer workload".

 

[Novuake]

Been waiting a while for this review. Glad to see Crucial sticking to the quality NAND.

 

[hannibal]

Not cheap but affordable indeed. And also as fast as high end ssd were some years ago.
Don't wanna see laptops with normal HD any more. In desktop normal HD still has superior price advantage.